Abstract:
Traditional electric power systems have several challenges in maintaining their reliability and being able to meet the demand of the consumers at peak hours. Additionally, environmental concerns may arise from several physical limitations in the network that would increase gas emission besides adding extra generation costs. With the advancements in the field of communications amalgamating in the power network, smart grids enable electric consumers to take part in changing the load profile through demand response (DR) programs to help overcome such challenges.
In some DR programs where the network’s operators inform the consumers about the updated prices, predicting the change of the consumption pattern that will occur becomes arduous. Especially with the variety of electrical loads and their applications like the residential and industrial consumers and their different sensitivity to prices.
For optimal scheduling of generation units, this thesis presents a novel method for the operator to predict market prices and electrical loads under real-time pricing (RTP) DR program in a microgrid. Inspired by the Stackelberg game, the proposed model represents the interaction between the operator and the consumers. The model establishes simulated trading between the network’s operator (leader) optimizing the generation cost and offering market prices to the customers (followers) who optimize their behavior. The interaction is formulated as a one-leader, N-follower iterative game where the optimization problems are solved using deterministic global optimization techniques. The proposed model considers a detailed representation of the industrial and residential loads. Simulations are performed on several microgrid systems where results show a significant improvement in the projected retail prices and electrical loads.
Finally, this thesis also examines the impact of energy storage systems (ESS) on the operation of an industrial facility in real-time demand response programs. A model is developed to optimally manage the energy storage and operation of the industrial load. Additionally, an approach to the sizing of the ESS is proposed. Stochastic modeling of electricity prices based on historical data is used to this end. The optimization models were tested on a generic industrial unit. Results show the benefits of ESS in increasing profit and highlight the impact of its installation cost on its feasibility.
